mattscq

Hello,

So I'm new to all this and after riding in the city for a few years, decided to embark on building a bike. I found a titanium frame of unknown provenance (all markings have been stripped but it does have a couple serial numbers under the BB) and thought the price was too good to pass up.

I have since learned many things.

Anyways, one of the things I've learned is that the frame is out of alignment. When measured, one side is off by about 1mm and the other side about 3-4mm. The rear spacing is still more or less 130mm (+/-1 mm) but is biased to one side a bit. I was told to talk to a local framebuilder (I'm in New York City) but they basically told me they were busy for the next 6 months, told me to call back way later and hung up on me. They did say they could fix it though. I took it around to a couple of shops and one seemed enthusiastic enough to try.

I've found there isn't a lot of info on cold-setting Ti and the few threads there are seem to be people wanting to modify their bikes to take different hub widths. I just want to straighten mine out.

Any ideas or suggestions? My basic research basically seems to suggest that it's possible but hard. You just need a lot more force and be more careful. Ti basically just has a higher yield point and its failure point isn't far off but as a metal, it basically goes through elasticity, deformation, then failure.

Are there additional things to consider when trying to bend Ti?

Thanks!

unterhausen

I would definitely consider not bending it at all.

There was a thread here last year where this was discussed at length. Some Ti is bendable, some isn't. Not sure how to find that thread though

mattscq

When a wheel was mounted to it, it was biased to one side by a noticeable amount and while there wasn't a tire on the rim we mounted, it seemed like there was a chance that a bump could momentarily get a wheel to rub up against the seatstays. The guy that was helping me said it wasn't unrideable but wouldn't stand by it himself. It did give a bit of a wobble.

Perhaps try to bend a little and see? I suppose I could live without perfect alignment but getting it a couple mm closer could make a huge difference.

I've also read somewhere that Ti has the potential to set back after you leave it for a few days, but that doesn't make sense to me. Once you take it past the elastic point, doesn't it just deform permanently?

Nessism

Please search for an answer. There has been threads on this already. Bottom line, it should be possible to straighten your frame, at least to a point.

Edit: a quick search turned up this...https://www.bikeforums.net/framebuil...-ti-frame.html

CliffordK

You could, of course, dish the wheel to be centered on the frame. A bit of a kluge, but it will work as long as the dropouts are reasonably square.

There is a repair shop in Portland that might be able to take on the task, and/or give you some advice.

https://www.ticycles.com/

It depends a bit on how much you wish to invest in the frame.

unterhausen

the numbers you describe don't really match up with a near rub. A rear triangle can be off a lot more than that without affecting the clearance at the tire. That suggests that the dropouts aren't in the right place along the stays. I would consider filing them or (and?) possibly adding some JB Weld if that would fix it.

Just to put it in perspective, a major American bike manufacturer told my LBS that their manufacturing tolerance is 5mm on dropout side-to-side alignment.

mattscq

Interesting. Somebody had filed a bit of the rear dropout before already (quite a noticeable amount really, a 2-3mm) and I wonder if that was because it was even more misaligned or when we test fitted the wheel, the guy wasn't paying attention to the weird mod. He mentioned dishing the wheel off axis but said that it would make for a wobbly ride and wasn't the safest thing to do.

I'm surprised a couple mm is causing so much grief too. Is perfect alignment really just an OCD thing?

ridelikeaturtle

I was kinda thinking the same thing, +/- a few mms doesn't make that much difference, does it? Can you not adjust the rear wheel in the dropouts so that difference is eliminated?

Also, I'd have thought there'd be too much spring in a titanium frame for it to be anywhere near easy to cold-set any better than where you're already at.

unterhausen

A couple of experts chimed in on the thread linked to above, and they said it's doable but not easy. With any ductile metal, you have to over-bend to compensate for springback.

In general, alignment does cause problems with a lot of things. Side to side is one of the least important though. It's also one of the easiest things for customers to measure. At least they think it is, not taking into account the lack of good references. I'm not sure how the OP measured, but the likelihood that it's the same as the builder measured is extremely low. The standard string test is a remarkably bad way to do it, as is the often used Park Frame Alignment Gauge-1 (abbreviation is censored here due to our diversity policy). The assumptions that the tubes are straight or the bb face is straight are generally not true post-weld. And also pre-weld.

I'm pretty sure that on the OP's frame, there was some mistake holding the dropouts in place. Dropouts slipped in the fixture or one of the stays moved or something like that. It's quite possible that fixing the side-to-side will make the rub worse. But 1mm on the one side is nothing, so I don't see any reason to mess with it. That's well within measurement error.

Trakhak

I'd definitely ask an experienced frame builder to look at it before putting money, time, and effort into fixing it, if, indeed, it needs fixing. But first, I'd stick a brand new, correctly dished wheel (without a tire mounted) in the frame and check the clearances.

Also, the filing that someone already did might have represented an attempt to fix a (possibly nonexistent) problem and might have thus created a problem.

The initial measurements might have been done incorrectly; in fact, given that you're "new at this" and that titanium frames have to be built to very close tolerances because they're so difficult to cold set, I'd put money on the measurements being off, especially if you and everyone you've shown the frame to have been using the string test or other iffy measuring techniques.

Example: a guy who worked at a shop that I managed (and who had just that week learned about measuring frame alignment) told me that a Marinoni custom-built frame that we'd ordered for a customer was out of alignment. I took his word for it and sent it back to Marinoni. I got a call from Marioni's headquarters a week later saying that (i) the frame measured as perfectly aligned on their alignment table and (ii) they were cutting us off as a Marinoni dealership.

mattscq

Thanks for all of your replies.


It was measured by two different people from different shops using what looked like a Parktool alignment gauge and they came to similar conclusions: about 4mm off on one side which I guess would mean ideally, you'd need to move both sides 2mm to re-center it and keep the 130mm spacing.


I did some quick math and geometrically, the rear wheel would be about .5 deg off. All of this seems minor but I keep getting people telling me this is not ideal or that I personally could but they wouldn't build it/stand by it etc.


I could probably get the frame back from the shop and try to find a new wheel to test it. I know Ti is hard to set. I don't mind letting this guy try to bend it. 2mm doesn't seem like an insane amount to move.


Thinking aloud:
If my rear dropout spacing is still 130 +/- 1mm, then it would seem like the whole rear chain/seatstay assembly got turned. If there's any wheel rubbing, that's probably due to misaligned dropouts for axle/wheel. I guess my bigger concern should be if the front and rear wheels (and the frame) will be parallel, but you're all saying that half a degree or a few mm shouldn't be noticeable?

unterhausen

The part I'm not following is that one side is off 4mm and the other off 1mm. That means that the drops should be noticeably too narrow or too wide. And also, it will spring when you put a wheel in, so the difference will be averaged. So the alignment should be better with a wheel in. I would say that most bike mechanics really don't know much about this subject. In any event, the Park tool doesn't measure anything worthwhile as far as tire clearance goes. Or at least it would have to be off a lot more than 4mm to make much of a difference.

One other thing that could be off is the position of the chainstays on the bb. Some builders don't hold this too well. I would look at that before doing much grinding on the drops.

mattscq

I think when they measured one side with the Park tool, it was off by 4-5mm, but the spacing was correct at 130. I would imply that that both stays would have to move over 2mm to be centered and symmetrical with the frame. It could also mean there's more to the frame that's misaligned?

I probably should get them to mount a new wheelset on it and see how it rolls/tracks. I don't really need it to be perfectly aligned as some sort of intellectual pursuit. I just want it to be safe and enjoyable to ride. It's just two different people told me it was 4mm off and an issue so I became worried.

mikeread

I have a Titanium frame that needs aligning, the wheel is not centered between the chainstays. I dont think it was always like this and I suspect it might have happened when I had a big crash a couple of years ago.

It must have been out of alignment for a long time before I noticed so it might be you can just build up your frame as is and it will be fine. Since building my own frame I tend to notice frame alignment and there are several riders in my club whose rear wheel is way off and I dont think any of them realise or notice! - expensive bikes too!

I do have some experience of bending grade 5 titanium (not with bike frames or tubes however) Heating to a dull red makes it much easier and will not affect the integrity of the material like it might steel, this is probably not necessary for the small bends we are talking about, but might be worth trying if you fail to manage it cold.

If you do try and bend your frame I would suggest having it clamped rigidly at the bottom bracket and have some method to control how far you are bending. Increase the bend distance a little at a time until the stay springs back to the desired position (measure between drop outs) Then do the other stay. Adding this level of control should prevent over bending.

One thing that has been puzzling me a bit and maybe the experienced builders here might be able to explain: If you have bent the rear triangle slightly and the wheel is now closer to the left hand chainstay, which way do you need to bend the stays to re-centre the wheel? Moving the drop outs to the right seems the obvious way at first but since they will be moving in an arc, the RH chainstay will be getting shorter and the LH longer which will rotate the wheel rim towards the LH chainstay making matters worse.

fietsbob

Out here I'd Consult with Dave Levy's Ti Cycles near Portland Oregon..

Butch Boucher

I peak in here once in a while and I'll chime in, I spent 21 years at Moots and aligned a lot of frames and have had to tune in fixtures to get a proper result through the frame building process. Keep in mind that moving the rear triangle around can be a slippery slope. The fact the dropouts have been filed indicates the builder knew something was off. If you move the dropouts a bunch to one side the faces will no longer be parallel and the alignment between the seatstays can be effected.

As far as moving the rear triangle if the chainstays are made out of thin larger diameter material you can buckle them. I've been around this. If they are smaller diameter they are probably a thicker and can be easier to move. At Moots we used US made CWSR seamless grade 9 tubing was a challenge to move much if it was out much. Building it straight in the first place is key. If it has been moved by being damaged it can be very difficult to get back.

I would find a builder familiar with Ti to do the alignment. They can check the entire frame for you and let you know the best way to get the best ride out of what you have. Not sure if Drew @ Engin down in the Philly area would do it, he has the skills, or maybe Steven @ Bilenky? Living in the mountains we don't think a thing about a 2 hour drive to get something done and they have good experience if they will do it.

unterhausen

It's funny, I don't remember what Drew has for alignment checks. Maybe it's because I was starstruck by everything else he has in his shop. I don't usually think of him when it comes to repairs, but if he would do it, he is a good choice. Bilenky is a good option for things like this, but I have heard they are backed up right now. So it will probably be a while.

mattscq

Thanks for all the comments and suggestions. I really appreciate it.

An update:

So a few days ago (after having been turned away by a couple local framebuilders: one didn't work in Ti and the other was so busy he told me to call back in 6 months), I left it with the local bike shop that seemed enthusiastic about taking a crack at it. Today they called and told me that they managed to move the non-drive side seat/chainstays without too much difficulty but the drive side was much less flexible. They've taken a break until tomorrow and they were going to try to investigate why that might be the case.

I'm frankly rather surprised it was moved with not too much effort (I got the sense that Ti would be a monumental pain to bend) but also how one side would move more easily than the other. Should I be worried?

Andrew R Stewart 

The first thought is how was the LBS securing the frame during the bending. Next is what is their lever arm. Depending on their set up it might be easier for them to push in one direction than the other. Andy

unterhausen

really hard to say from a distance, but this is the second best time to find out it's cracked. First being before you bought it

squirtdad

op post a bit more and put up some pics